[PATCH] D69395: [AMDGPU] Fix mfma scheduling crash

Stanislav Mekhanoshin via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Thu Oct 24 11:08:58 PDT 2019


This revision was automatically updated to reflect the committed changes.
Closed by commit rG3c8e055187d8: [AMDGPU] Fix mfma scheduling crash (authored by rampitec).

Changed prior to commit:
  https://reviews.llvm.org/D69395?vs=226291&id=226303#toc

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D69395/new/

https://reviews.llvm.org/D69395

Files:
  llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
  llvm/test/CodeGen/AMDGPU/mfma-loop.ll


Index: llvm/test/CodeGen/AMDGPU/mfma-loop.ll
===================================================================
--- llvm/test/CodeGen/AMDGPU/mfma-loop.ll
+++ llvm/test/CodeGen/AMDGPU/mfma-loop.ll
@@ -76,5 +76,39 @@
   ret void
 }
 
+; GCN-LABEL: {{^}}test_mfma_loop_mfma_forward_init:
+
+; GCN-COUNT32: v_accvgpr_write_b32 a{{[0-9]+}}, 0
+; GCN: v_mfma_f32_32x32x1f32
+; GCN-NOT: v_accvgpr
+
+; GCN: [[LOOP:BB[0-9_]+]]:
+; GCN-NOT: v_accvgpr
+; GCN: v_mfma_f32_32x32x1f32
+; GCN-NOT: v_accvgpr
+; GCN: s_cbranch_scc1 [[LOOP]]
+
+; GCN-COUNT32: v_accvgpr_read_b32
+define amdgpu_kernel void @test_mfma_loop_mfma_forward_init(<32 x float> addrspace(1)* %arg) {
+entry:
+  %tid = call i32 @llvm.amdgcn.workitem.id.x()
+  %init = bitcast i32 %tid to float
+  %mai.0 = tail call <32 x float> @llvm.amdgcn.mfma.f32.32x32x1f32(float 1.0, float 2.0, <32 x float> zeroinitializer, i32 0, i32 0, i32 0)
+
+  br label %for.cond.preheader
+
+for.cond.preheader:
+  %phi = phi <32 x float> [ %mai.0, %entry ], [ %mai.1, %for.cond.preheader ]
+  %c = phi i32 [ 0, %entry ], [ %inc, %for.cond.preheader ]
+  %mai.1 = tail call <32 x float> @llvm.amdgcn.mfma.f32.32x32x1f32(float 1.0, float 2.0, <32 x float> %phi, i32 0, i32 0, i32 0)
+  %inc = add nuw nsw i32 %c, 1
+  %cc = icmp eq i32 %inc, 16
+  br i1 %cc, label %exit, label %for.cond.preheader
+
+exit:
+  store <32 x float> %mai.1, <32 x float> addrspace(1)* %arg
+  ret void
+}
+
 declare <32 x float> @llvm.amdgcn.mfma.f32.32x32x1f32(float, float, <32 x float>, i32, i32, i32)
 declare i32 @llvm.amdgcn.workitem.id.x()
Index: llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
===================================================================
--- llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
+++ llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
@@ -773,6 +773,11 @@
     return MI && TII->isSALU(*MI) && !MI->isTerminator();
   }
 
+  bool isVALU(const SUnit *SU) const {
+    const MachineInstr *MI = SU->getInstr();
+    return MI && TII->isVALU(*MI);
+  }
+
   bool canAddEdge(const SUnit *Succ, const SUnit *Pred) const {
     if (Pred->NodeNum < Succ->NodeNum)
       return true;
@@ -821,7 +826,7 @@
 
       for (SDep &SI : From->Succs) {
         SUnit *SUv = SI.getSUnit();
-        if (SUv != From && TII->isVALU(*SUv->getInstr()) && canAddEdge(SUv, SU))
+        if (SUv != From && isVALU(SUv) && canAddEdge(SUv, SU))
           SUv->addPred(SDep(SU, SDep::Artificial), false);
       }
 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D69395.226303.patch
Type: text/x-patch
Size: 2431 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191024/b7c9577b/attachment-0001.bin>


More information about the llvm-commits mailing list